Acquisition and understanding of process knowledge using problem solving methods

En savoir plus sur le livre

The development of knowledge-based systems relies on knowledge engineers (KEs) and subject matter experts (SMEs). A crucial step in this process is transferring knowledge from SMEs to formal, machine-readable formats that enable reasoning. This task is often costly and prone to errors. To overcome the knowledge acquisition bottleneck, it is essential to empower SMEs to create these knowledge representations independently, particularly for complex knowledge types like processes. Analysis across various application domains reveals that process knowledge is frequently encountered, and its complexity necessitates specialized methods for SMEs to represent it computationally. Moreover, the vast data generated from process executions in knowledge-intensive fields such as Biology and Astronomy demands analytical tools with high abstraction capabilities to assist SMEs in analyzing these processes. This work introduces methods and tools that facilitate SMEs in acquiring, formally representing, reasoning about process knowledge, and analyzing process executions through provenance. We highlight the use of Problem Solving Methods as key knowledge artifacts for process acquisition and analysis, serving both as formalizations of reasoning strategies for processes and as high-level, domain-independent abstractions that offer SMEs insights into process executions.